The Design Review - John Dunn, Consultant, Ambertec, P.E., P.C.

Every circuit design has some chance of benefitting from having a critical eye cast upon it by someone other than its creator. You think you know your work, you've put your heart and soul into it and so you have all kinds of confidence, but darned if there isn't that little blind spot somewhere that prevents you from noticing something.

A professor I once had put it quite nicely: "Familiarity breeds the illusion of comprehension."

Thus there was this particular project that I brought into a design review, hoping to get a careful and thoughtful evaluation of my work. The profundity of the review commentary was astonishing:

            "Well, you know you have to keep all of the critical leads as short as possible, be sure that

            you've bypassed your rail voltages properly, make sure that all of the parts tolerances are

            reasonable and you also have to make sure nothing will overheat." 

 Now comes the sarcasm:

            I don't know why they were so critical. The bypass capacitors were only a tenth of a microfarad

            which meant that having them six inches from their respective op-amps shouldn't have hurt

            anything, right?

 

            The summing junctions are virtual ground after all, so having those nodes stretched out across

            a few inches of the circuit board couldn't be a problem either, right?

 

            Parts tolerances are never a concern either. We all know that 1% resistors are really better than

            that. After all, they're precision parts, right?

 

            Also, what's all this business about temperature? Everybody knows you can run your ICs and

            transistors at higher than maximum rated temperatures because the semiconductor companies

            always build in a safety factor for that, right?

 

Back to the real world:

My design passed its review, but the reviewers weren't really helpful. The critical eye that I was really hoping for, that would really have ferreted out any unnoticed problem if such were there, was not there.

Instead, the review was merely a recitation of the standard generalized litany of good practices and as non-controversial as anything in this business can get, If I had missed something (I hadn't as it turned out), I don't think it would have been caught.

In the 1950's, there were commonly used expressions which I detested like "Don't rock the boat.", "Go with the flow.", "Ride with the tide.", "Don't be such a d.....ed. perfectionist." and the Madison Avenue advertising business line "Let's run it up the flagpole and see who salutes." 

All of these phrases indicated to me abdications of responsibility in whatever situation was at hand and complete abdication was the hallmark of that design review.

Of course, you and I wouldn't do that. Of course not.

Comments

[Graham Holmes]

It all sounds very familiar, John. You don't say whether the reviewers were sticking to neutral territory as a result of some political arm-twisting, or whether they were just unqualified to offer criticism. In my experience, it's usually politics. I was once told not to raise any problems at a design review because the project couldn't be allowed to fail. (In fact, the review was just a box-ticking exercise, because the product was already being deployed.) Heard that before? Of course - remember what happened to Morton Thiokol and NASA when politics stifled scientific criticism. As a designer/developer of electronics and software, I strongly believe in the review process. But if the conclusion has been reached before the review has even been planned, then you may as well take the day off. Sadly, it happens all too often.

[Cor van de Water]

I guess it depends on the leadership of the manager.
As long as I was managing the hardware devt team, the engineers
had better review any design thoroughly, because I know from
experience that anything you catch early costs wayyy less to
fix than letting it slide into the next phase of devt...
Eventually the customer is going to find it so you do your
darnest to avoid letting it go that far.
But that is just my humble / engineer-speaking opinion.

[Frank Walker]

Luckily, I haven't had the experiences you have. Although I have met my fair share of incompetent engineers... and even though I hate to generalize, most of these individuals came from schools which are extremely prestigious. Simply put: They were advanced in their career past their capabilities because they had the resume. I took the design review from a (IMHO Junior) Senior engineer as a chance to have dialog and possibly share some knowledge with the person evaluating my design.

All of us came out of school knowing eigenvalues, transforms, and the fundamentals of engineering enough to pass a test. Over the next 5-10 years we are molded into engineers, usually picked up by a mentor who is up for the challenge of growing some talent. (I myself was fortunate in this sense.)

For the most part, design reviews generated reasonable feedback and helped to bring the project together.

We all find individuals who are incompetent, lazy or simply just don't care. From my experience these people typically usually end up moving down the road to something they are actually fit for... usually managment! HA!

[Jeff Furman]

On occasion, I've been handed a stack of review drawings just before the review meeting; I know I can't do that review justice, regardless of my experience. Perhaps my favorite 'hot button' issue will yield some insight, but that's rolling the dice. That initial pause in a review while participants try to 'do their homework' in place is not a good sign.